Make an image classification app using React and Tensorflow JS
Vložit
- čas přidán 4. 09. 2024
- In this video, we will learn how to make a simple Image Classification / Identification app using React and Tensorflow JS. Tensorflow JS is a machine learning library for Javascript. For this tutorial, there is no machine learning knowledge required.
We are going to use the pre-trained machine learning model from Tensorflow JS called mobilenet. This model is specifically used for image classification. This is a very simple app that uses tensorflow model, and tries to identify the content of the image.
So if we use an image of animals like cats, dogs or other simple things, The Tensorflow will work behind the scenes to identify the content for this image, and will display the results in 3 guesses.
Link to the finished product: imageid.netlif...
Link to the Git Repository: bitbucket.org/...
If you find this video helpful or informative, please like, share and comment on this video, and Please subscribe to my channel for more videos like this.
Thank you for watching.
Buy me a Coffee: paypal.me/thou...
Because you are fluent in this and understand your 'industry', you are brilliant in transferring the knowledge in a highly excellent and organised way. The creative is creative. Thank you.
Thank you for watching :)
You deserve more likes. This is a masterclass. Thank you very much.
Thank you for your comment and Thanks for watching :) Cheers :)
Very organised tutorial... Do more of these..
Thank you for your comment and Thanks for watching :) Cheers :)
Thank you for showing me how to use tensorflow in a react app
Thank you for your comment and Thanks for watching :) Cheers :)
thank you for this tutorial! I've learned so many things!
Thank you for your comment and Thanks for watching :) Cheers :)
A Shame you stopped making videos! Awesome tutorial and explanation ! thanks for sharing
Awsome video bro you deserve way much more likes then you have crisp clear explaination fluent english and a very big brain keep it up you will reach at the top bro 🚀🚀🚀🚀🚀🔥🔥🔥🔥🔥🔥
updated dependencies are available, and there is a lot of error downloading the old ones. how can i solve that?
i mean tensorflow js
Thanks a lot for this awesome content 😄
Great stuff
Awesome! Subscribed fella, here!
That was really well done 👍
Thank you!
Hi,
Hope you are doing well. Can You make a video on how to use custom trained model in react native app???
Furthermore if i want to use multiple models in one react native app how to do it?
thanks
why mines error says "No backend found in registry."?
Hi is it possible to limit the classification like example when I want flowers only? Using this?
The other sources use android as their IDE which my laptop can't handle because its only 2 gb ram
Thanks for this btw I have hope now and ideas for my upcoming final year school project
Can u tell the windows command as I was using windows os
What is the model used in this project and does it use progressive web app (pwa)?
Not working for me. model not loading
classify is showing error instead of array
Can you upload how to coding for react + tensorflow.js object detection?
Thank you so much for such an awesome video. Can you help me with one thing?
How can I use my custom model in this app?
If you can help me I would be really grateful.
Thank you in advance
Is the css automatically aligned??
what about backend
can we use our own model instead of tensorflow-models if yes please what to change
const results = await model.classify(imageRef.current)
this line is showing error
i have the same issue, did you solve it?
downgrade tensorflow/tfjs-converter and tensorflow/tfjs-cors module to 1.2.1 version to use the model
It's show me this error.
pls help
Module not found: Error: Can't resolve '@tensorflow-models/mobilenet' in 'E:\potatodisease\my-app\src'
downgrade tensorflow/tfjs-converter and tensorflow/tfjs-cors module to 1.2.1 version to use the model
@@viralgupta7636 its still showing errors
@@vedikatyagi3269 did you solve it?
Can we have a tutorial on how to deploy this?
the dependencies does not support
models is null even using old version
how to change animal to plant detection ??
Ideally it should do that automatically, but honestly I haven't tried it :). The library I am using a pre-trained machine learning model which is a third party library and is trained, maintained and constantly updated by Google. So I can say over a period of time, it will become sophisticated enough to identify and classify different plants, flowers and many other objects :) Thanks for Watching this video. Cheers :)
Is this code works??
source code?
Very interesting. At this time your code is failing to run, any update ??
I think with the latest updates a lot might have changed. But If you try installing the packages with the same version that I am using in the video, then it should work. Let me know how it goes. :)
@@thoughtspan No I didn't try with your version, I just wanted to adapt your code to new versions but I failed.
@@thoughtspan i tried with same packages, and i still seeing an error in "const results = await model.classify(imageRef.current)" console says " Uncaught (in promise) TypeError: Cannot read properties of null (reading 'classify')", idk what to do :(
downgrade tensorflow/tfjs-converter and tensorflow/tfjs-cors module to 1.2.1 version to use the model
how can i load custom model ?
You can try to use one of the pretrained models from this link: www.tensorflow.org/js/models
NOT A HOTDOG
Ha Ha :P I was thinking of the same scene from Silicon valley when I was making this video :) Thanks for watching :)
I just went down looking for this comment
Clicking on 'upload image' opens up my iphone camera but with a black screen. It doesnt capture anything nor does it give option to upload images from gallery
Which iPhone version did you use? I don't have an iPhone. I used my Android, and it worked well. I will test it on an iPhone too.